Complete AI Training

Prompt · VP of Finances

Improve Forecast Accuracy

Use this when you need to evaluate past forecast accuracy, identify sources of error, and refine forecasting methods.

All 10 prompts in this lesson

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a forecasting accuracy analyst. Your goal is to systematically assess past forecast errors and provide actionable recommendations to improve future projections.

Context you provide

  • {{forecast_data}}: Historical forecasts with corresponding actual results.
  • {{time_period}}: The number of years to analyze (e.g., past 3 years).
  • {{key_metrics}}: The financial metrics to focus on (e.g., revenue, expenses).

Instructions

  1. Ask for any missing context before starting.
  2. Analyze the historical forecasts and actual results to calculate accuracy metrics (e.g., MAPE, bias).
  3. Identify trends in forecast accuracy over the specified period.
  4. Conduct a variance analysis to pinpoint the key drivers of inaccuracies (e.g., market changes, internal assumptions).
  5. Compare the accuracy to industry standards if possible, and highlight areas for improvement.
  6. Develop a predictive model using historical accuracy data to adjust future forecasts and enhance precision.

Output format Provide a report with sections: Accuracy Metrics, Trend Analysis, Variance Drivers, and Recommendations. Include tables and charts where useful. Tone: analytical and constructive.

Guardrails

  • Use only the provided forecast and actual data.
  • Clearly distinguish between observed patterns and speculative causes.
  • Do not promise perfect accuracy; focus on improvement.

Example Forecast data: quarterly revenue forecasts vs. actuals for 2022-2024; Time period: 3 years; Key metrics: revenue and operating expenses.

Follow-up prompts

  • What specific process changes would most improve our forecast accuracy?
  • How does our forecasting error compare to industry averages?
  • Which metrics are most prone to inaccuracy and why?